MCPcopy Create free account

hub / github.com/brycedrennan/imaginAIry / functions

Functions2,992 in github.com/brycedrennan/imaginAIry

⨍ Functions 2,992◇ Types & classes 731↳ Endpoints 37
↓ 1 callersFunctionload_checkpoint
(model, url_or_filename)
imaginairy/vendored/blip/blip_nlvr.py:88
↓ 1 callersFunctionload_checkpoint
(fpath, model)
imaginairy/vendored/imaginairy_normal_map/model.py:76
↓ 1 callersFunctionload_controlnet
(control_weights_location, half_mode)
imaginairy/utils/model_manager.py:398
↓ 1 callersFunctionload_flux_models
()
imaginairy/api/generate_flux.py:12
↓ 1 callersMethodload_from_pretrained
(self, name="vgg_lpips")
imaginairy/modules/sgm/autoencoding/lpips/loss/lpips.py:28
↓ 1 callersFunctionload_from_safetensors
(path: Path | str, device: Device | str = "cpu")
imaginairy/vendored/refiners/fluxion/utils.py:176
↓ 1 callersFunctionload_gitignore_spec_at_path
(path)
imaginairy/utils/gitignore.py:55
↓ 1 callersFunctionload_image
Converts a LazyLoadingImage or PIL Image into a PyTorch tensor.
imaginairy/enhancers/upscale.py:99
↓ 1 callersMethodload_image_from_base64
(image_str: str)
imaginairy/schema.py:201
↓ 1 callersFunctionload_midas
(model_type="dpt_hybrid")
imaginairy/modules/midas/api.py:311
↓ 1 callersFunctionload_model
Load the specified network. Args: device (device): the torch device used model_path (str): path to saved model model_type
imaginairy/modules/midas/api.py:98
↓ 1 callersFunctionload_model
( config: str, device: str, num_frames: int, num_steps: int, weights_url: str )
imaginairy/api/video_sample.py:398
↓ 1 callersMethodload_model
(self, path, version: float)
imaginairy/enhancers/video_interpolation/rife/RIFE_HDv3.py:14
↓ 1 callersFunctionload_model_from_config
(config, weights_location, half_mode=False)
imaginairy/utils/model_manager.py:105
↓ 1 callersMethodload_pretrained_ckpt
(self)
imaginairy/vendored/facexlib/matting/backbone.py:22
↓ 1 callersFunctionload_sd15_pipeline
Load a diffusion model. Weights location may also be shortcut name, e.g. "SD-1.5"
imaginairy/utils/model_manager.py:262
↓ 1 callersFunctionload_sdxl_compvis_weights
(url)
imaginairy/utils/model_manager.py:794
↓ 1 callersFunctionload_sdxl_pipeline
(base_url, device=None, for_inpainting=False)
imaginairy/utils/model_manager.py:660
↓ 1 callersFunctionload_sdxl_pipeline_from_compvis_weights
( base_url: str, for_inpainting=False, device=None, dtype=torch.float16 )
imaginairy/utils/model_manager.py:622
↓ 1 callersFunctionload_sdxl_pipeline_from_diffusers_weights
( base_url: str, for_inpainting=False, device=None, dtype=torch.float16 )
imaginairy/utils/model_manager.py:551
↓ 1 callersMethodload_text_encoder
(self, loras: dict[str, Lora], /)
imaginairy/vendored/refiners/foundationals/latent_diffusion/lora.py:54
↓ 1 callersFunctionload_txt_file
Load data or string from txt file.
imaginairy/vendored/facexlib/alignment/convert_98_to_68_landmarks.py:4
↓ 1 callersMethodload_unet
(self, loras: dict[str, Lora], /)
imaginairy/vendored/refiners/foundationals/latent_diffusion/lora.py:58
↓ 1 callersMethodload_untranslated_weights
( self, source_path: str, device: Device | str = "cpu" )
imaginairy/weight_management/translation.py:25
↓ 1 callersMethodload_untranslated_weights
( self, text_encoder_l_weights_path: str, text_encoder_g_weights_path: str, de
imaginairy/weight_management/translators.py:74
↓ 1 callersMethodload_weights
(self, key_tensor: Tensor, value_tensor: Tensor)
imaginairy/vendored/refiners/foundationals/latent_diffusion/image_prompt.py:325
↓ 1 callersMethodlog_conditioning
(self, conditioning, description)
imaginairy/utils/log_utils.py:272
↓ 1 callersMethodlog_images
( self, batch, N=8, n_row=2, *, sample=True, return_keys=None, **kwargs )
imaginairy/modules/diffusion/ddpm.py:667
↓ 1 callersMethodlog_latents
(self, latents, description)
imaginairy/utils/log_utils.py:281
↓ 1 callersFunctionlog_txt_as_img
(wh, xc, size=10)
imaginairy/modules/diffusion/ddpm.py:46
↓ 1 callersMethodlogging_dict
Return a dict of the object but with binary data replaced with reprs.
imaginairy/schema.py:801
↓ 1 callersFunctionlookup_top_adapter
Lookup and return last adapter in parents tree (or target if none).
imaginairy/vendored/refiners/fluxion/adapters/adapter.py:88
↓ 1 callersMethodlora_layers
( self, device: Device | str | None = None, dtype: DType | None = None )
imaginairy/vendored/refiners/fluxion/adapters/lora.py:30
↓ 1 callersFunctionmain
Script to convert the controlnet weights into diffs that are ready to be applied to any s1.5 weights.
scripts/controlnet_convert.py:12
↓ 1 callersFunctionmain
()
docs/examples/generate_doc_examples.py:5
↓ 1 callersFunctionmake_bbox_head
(fpn_num=3, inchannels=64, anchor_num=2)
imaginairy/vendored/facexlib/detection/retinaface_net.py:185
↓ 1 callersFunctionmake_class_head
(fpn_num=3, inchannels=64, anchor_num=2)
imaginairy/vendored/facexlib/detection/retinaface_net.py:178
↓ 1 callersMethodmake_clothing_segmentation_diagram
(self)
imaginairy/img_processors/segformer_b2_clothes.py:89
↓ 1 callersFunctionmake_clothing_segmentation_map
(image_pil)
imaginairy/img_processors/segformer_b2_clothes.py:142
↓ 1 callersMethodmake_cond_schedule
( self, )
imaginairy/modules/diffusion/ddpm.py:825
↓ 1 callersMethodmake_cpu_space
(self, bytes_to_free)
imaginairy/utils/model_cache.py:179
↓ 1 callersFunctionmake_ddim_sampling_parameters
(alphacums, ddim_timesteps, eta)
imaginairy/modules/diffusion/util.py:94
↓ 1 callersFunctionmake_ddim_timesteps
( ddim_discr_method, num_ddim_timesteps, num_ddpm_timesteps, )
imaginairy/modules/diffusion/util.py:71
↓ 1 callersFunctionmake_int_box
(box: torch.Tensor)
imaginairy/img_processors/densepose.py:566
↓ 1 callersFunctionmake_landmark_head
(fpn_num=3, inchannels=64, anchor_num=2)
imaginairy/vendored/facexlib/detection/retinaface_net.py:192
↓ 1 callersFunctionmake_layer
Make layers by stacking the same blocks. Args: basic_block (nn.module): nn.module class for basic block. num_basic_block (int): n
imaginairy/vendored/basicsr/arch_util.py:46
↓ 1 callersFunctionmake_mp4_animation
(imgs, outpath, frame_duration_ms=50, fps=30, codec="mp4v")
imaginairy/utils/animations.py:140
↓ 1 callersFunctionmake_safe_filename
(input_string)
imaginairy/api/video_sample.py:465
↓ 1 callersMethodmake_simple_clothing_segmentation_diagram
(self)
imaginairy/img_processors/segformer_b2_clothes.py:99
↓ 1 callersFunctionmake_slideshow_animation
( imgs, outpath, image_pause_ms=1000, )
imaginairy/utils/animations.py:89
↓ 1 callersFunctionmake_txts
()
scripts/prep_vocab_lists.py:51
↓ 1 callersMethodmap_state_dicts
Find a mapping between the source and target models' state_dicts. - `source_args`: The arguments to pass to the source model it can
imaginairy/vendored/refiners/fluxion/model_converter.py:253
↓ 1 callersMethodmapping
(self)
imaginairy/weight_management/conversion.py:47
↓ 1 callersFunctionmask_tile
(tile, overlap, std_overlap, side="bottom")
imaginairy/utils/feather_tile.py:11
↓ 1 callersFunctionmask_to_bbox
(mask_img_t)
imaginairy/img_processors/densepose.py:108
↓ 1 callersFunctionmatch_brightness
(target_img, source_img)
imaginairy/api/colorize.py:78
↓ 1 callersFunctionmeasure_perplexity
( predicted_indices: torch.Tensor, num_centroids: int )
imaginairy/modules/sgm/autoencoding/regularizers/base.py:31
↓ 1 callersMethodmeshgrid
(self, h, w)
imaginairy/modules/diffusion/ddpm.py:921
↓ 1 callersMethodmetadata_dict
(self)
imaginairy/schema.py:889
↓ 1 callersFunctionmod_get_invoke
Get the Cmd main method from the click command :param command: The click Command object :return: the do_* method for Cmd :rtype: func
imaginairy/cli/clickshell_mod.py:18
↓ 1 callersMethodmode
(self)
imaginairy/modules/sgm/distributions/distributions.py:73
↓ 1 callersFunctionmonkeypatch_safety_cosine_distance
()
imaginairy/utils/safety.py:130
↓ 1 callersFunctionmonkeypatch_sd1controlnetadapter
Another horrible thing. I needed to be able to cache the controlnet objects so I wouldn't be making new ones on every image generation.
imaginairy/modules/refiners_sd.py:836
↓ 1 callersFunctionmonkeypatch_self_attention_guidance
()
imaginairy/modules/refiners_sd.py:874
↓ 1 callersFunctionmsssim
( img1, img2, window_size=11, size_average=True, val_range=None, normalize=False )
imaginairy/enhancers/video_interpolation/rife/msssim.py:214
↓ 1 callersMethodmultistep_dpm_solver_second_order_update
(self, x: Tensor, step: int)
imaginairy/vendored/refiners/foundationals/latent_diffusion/schedulers/dpm_solver.py:70
↓ 1 callersMethodnoise_an_image
(self, init_latent, t, schedule, noise=None)
imaginairy/samplers/ddim.py:223
↓ 1 callersMethodnoise_an_image
(self, init_latent, t, schedule, noise=None)
imaginairy/samplers/plms.py:232
↓ 1 callersFunctionopenpose_model
()
imaginairy/img_processors/openpose.py:506
↓ 1 callersMethodp_losses
(self, x_start, t, noise=None)
imaginairy/modules/diffusion/ddpm.py:555
↓ 1 callersMethodp_losses
(self, x_start, cond, t, noise=None)
imaginairy/modules/diffusion/ddpm.py:1295
↓ 1 callersMethodp_mean_variance
(self, x, t, clip_denoised: bool)
imaginairy/modules/diffusion/ddpm.py:469
↓ 1 callersMethodp_mean_variance
( self, x, c, t, clip_denoised: bool, return_codebook_ids=Fals
imaginairy/modules/diffusion/ddpm.py:1331
↓ 1 callersMethodp_sample
(self, x, t, clip_denoised=True, repeat_noise=False)
imaginairy/modules/diffusion/ddpm.py:484
↓ 1 callersMethodp_sample_ddim
( self, noisy_latent, neutral_conditioning, positive_conditioning, gui
imaginairy/samplers/ddim.py:117
↓ 1 callersMethodp_sample_loop
(self, shape, return_intermediates=False)
imaginairy/modules/diffusion/ddpm.py:495
↓ 1 callersMethodp_sample_plms
( self, noisy_latent, neutral_conditioning, positive_conditioning, gui
imaginairy/samplers/plms.py:139
↓ 1 callersFunctionpad_bbox
(bbox, max_height, max_width, pad=1)
imaginairy/img_processors/densepose.py:119
↓ 1 callersFunctionpad_right_down_corner
(img, stride, padValue)
imaginairy/img_processors/openpose.py:18
↓ 1 callersFunctionparse_csv_line
(line)
imaginairy/utils/prompt_schedules.py:67
↓ 1 callersFunctionparse_csv_line
(line)
imaginairy/cli/arg_schedule.py:63
↓ 1 callersFunctionparse_html_tag
Parses a single HTML tag and returns a dictionary with the tag name and its attributes. Args: html_tag (str): A string representing the
imaginairy/utils/spaced_kv_parser.py:51
↓ 1 callersFunctionparse_schedule_str
Parse a schedule string into a list of values.
imaginairy/cli/arg_schedule.py:42
↓ 1 callersFunctionparse_schedule_strs
Parse and validate input prompt schedules.
imaginairy/utils/prompt_schedules.py:33
↓ 1 callersFunctionparse_schedule_strs
Parse and validate input prompt schedules.
imaginairy/cli/arg_schedule.py:27
↓ 1 callersFunctionpaste_face_back
(img, face, inverse_affine)
imaginairy/vendored/facexlib/utils/face_utils.py:190
↓ 1 callersMethodpaste_faces_to_input_image
(self, save_path=None, upsample_img=None)
imaginairy/vendored/facexlib/utils/face_restoration_helper.py:281
↓ 1 callersFunctionpatch_download_function_in_facexlib_modules
Replaces the custom weights downloaded with the standard imaginairy one.
imaginairy/enhancers/face_restoration_codeformer.py:30
↓ 1 callersFunctionpath_is_ignored
(path: Path, gitignore_dict)
imaginairy/utils/gitignore.py:102
↓ 1 callersFunctionpillow_fit_image_within
( image: Image.Image, max_height=512, max_width=512, convert="RGB", snap_size=8 )
imaginairy/api/video_sample.py:441
↓ 1 callersFunctionpillow_img_to_torch_normal_map_input
(img)
imaginairy/vendored/imaginairy_normal_map/model.py:51
↓ 1 callersFunctionpoint_form
Convert prior_boxes to (xmin, ymin, xmax, ymax) representation for comparison to point form ground truth data. Args: boxes: (tensor)
imaginairy/vendored/facexlib/detection/retinaface_utils.py:50
↓ 1 callersMethodpoints_to_tensor
( self, foreground_points: Sequence[tuple[float, float]] | None = None, background_poi
imaginairy/vendored/refiners/foundationals/segment_anything/prompt_encoder.py:112
↓ 1 callersMethodpool
(self, x, text)
imaginairy/modules/sgm/encoders/modules.py:496
↓ 1 callersMethodpossible_correction_step
( self, euler_step, x, d, dt, next_sigma, denoiser, cond, uc )
imaginairy/modules/sgm/diffusionmodules/sampling.py:224
↓ 1 callersMethodpossibly_get_ucg_val
(self, embedder: AbstractEmbModel, batch: Dict)
imaginairy/modules/sgm/encoders/modules.py:131
↓ 1 callersMethodpossibly_quantize_c_noise
(self, c_noise: torch.Tensor)
imaginairy/modules/sgm/diffusionmodules/denoiser.py:25
↓ 1 callersMethodpossibly_quantize_sigma
(self, sigma: torch.Tensor)
imaginairy/modules/sgm/diffusionmodules/denoiser.py:22
↓ 1 callersMethodpostprocess_masks
(self, masks: Tensor, target_size: tuple[int, int], original_size: tuple[int, int])
imaginairy/vendored/refiners/foundationals/segment_anything/model.py:140
↓ 1 callersMethodpredict
Advances the state vector and returns the predicted bounding box estimate.
imaginairy/vendored/facexlib/tracking/kalman_tracker.py:91
← previousnext →1,001–1,100 of 2,992, ranked by callers